--------------------------- dentry_operations --------------------------
prototypes:
- int (*d_revalidate)(struct dentry *, int);
+ int (*d_revalidate)(struct dentry *, struct nameidata *);
int (*d_hash)(const struct dentry *, const struct inode *,
struct qstr *);
int (*d_compare)(const struct dentry *, const struct inode *,
char *(*d_dname)((struct dentry *dentry, char *buffer, int buflen);
locking rules:
- rename_lock ->d_lock may block
-d_revalidate: no no yes
-d_hash no no no
-d_compare: yes no no
-d_delete: no yes no
-d_release: no no yes
-d_iput: no no yes
-d_dname: no no no
+ rename_lock ->d_lock may block rcu-walk
+d_revalidate: no no yes (ref-walk) maybe
+d_hash no no no maybe
+d_compare: yes no no maybe
+d_delete: no yes no no
+d_release: no no yes no
+d_iput: no no yes no
+d_dname: no no no no
--------------------------- inode_operations ---------------------------
prototypes:
The cases where rcu-walk cannot continue are:
* NULL dentry (ie. any uncached path element)
* parent with d_inode->i_op->permission or ACLs
-* dentries with d_revalidate
* Following links
-In future patches, permission checks and d_revalidate become rcu-walk aware. It
-may be possible eventually to make following links rcu-walk aware.
+In future patches, permission checks become rcu-walk aware. It may be possible
+eventually to make following links rcu-walk aware.
Uncached path elements will always require dropping to ref-walk mode, at the
very least because i_mutex needs to be grabbed, and objects allocated.
INIT_LIST_HEAD(&inode->i_dentry);
must be done in the RCU callback.
+
+--
+[recommended]
+ vfs now tries to do path walking in "rcu-walk mode", which avoids
+atomic operations and scalability hazards on dentries and inodes (see
+Documentation/filesystems/path-walk.txt). d_hash and d_compare changes (above)
+are examples of the changes required to support this. For more complex
+filesystem callbacks, the vfs drops out of rcu-walk mode before the fs call, so
+no changes are required to the filesystem. However, this is costly and loses
+the benefits of rcu-walk mode. We will begin to add filesystem callbacks that
+are rcu-walk aware, shown below. Filesystems should take advantage of this
+where possible.
+
+--
+[mandatory]
+ d_revalidate is a callback that is made on every path element (if
+the filesystem provides it), which requires dropping out of rcu-walk mode. This
+may now be called in rcu-walk mode (nd->flags & LOOKUP_RCU). -ECHILD should be
+returned if the filesystem cannot handle rcu-walk. See
+Documentation/filesystems/vfs.txt for more details.
dcache. Most filesystems leave this as NULL, because all their
dentries in the dcache are valid
+ d_revalidate may be called in rcu-walk mode (nd->flags & LOOKUP_RCU).
+ If in rcu-walk mode, the filesystem must revalidate the dentry without
+ blocking or storing to the dentry, d_parent and d_inode should not be
+ used without care (because they can go NULL), instead nd->inode should
+ be used.
+
+ If a situation is encountered that rcu-walk cannot handle, return
+ -ECHILD and it will be called again in ref-walk mode.
+
d_hash: called when the VFS adds a dentry to the hash table. The first
dentry passed to d_hash is the parent directory that the name is
to be hashed into. The inode is the dentry's inode.
* yet completely filled in, and revalidate has to delay such
* lookups..
*/
-static int autofs_revalidate(struct dentry * dentry, struct nameidata *nd)
+static int autofs_revalidate(struct dentry *dentry, struct nameidata *nd)
{
struct inode * dir;
struct autofs_sb_info *sbi;
struct autofs_dir_ent *ent;
int res;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
lock_kernel();
dir = dentry->d_parent->d_inode;
sbi = autofs_sbi(dir->i_sb);
#include <linux/ctype.h>
#include <linux/net.h>
#include <linux/sched.h>
+#include <linux/namei.h>
#include "smb_fs.h"
#include "smb_mount.h"
* This is the callback when the dcache has a lookup hit.
*/
static int
-smb_lookup_validate(struct dentry * dentry, struct nameidata *nd)
+smb_lookup_validate(struct dentry *dentry, struct nameidata *nd)
{
- struct smb_sb_info *server = server_from_dentry(dentry);
- struct inode * inode = dentry->d_inode;
- unsigned long age = jiffies - dentry->d_time;
+ struct smb_sb_info *server;
+ struct inode *inode;
+ unsigned long age;
int valid;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ server = server_from_dentry(dentry);
+ inode = dentry->d_inode;
+ age = jiffies - dentry->d_time;
+
/*
* The default validation is based on dentry age:
* we believe in dentries for a few seconds. (But each
#include <linux/module.h>
#include <linux/init.h>
#include <linux/fs.h>
+#include <linux/namei.h>
#include <linux/pagemap.h>
#include <linux/ctype.h>
#include <linux/sched.h>
void *dir_version;
int ret;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
vnode = AFS_FS_I(dentry->d_inode);
if (dentry->d_inode)
*/
static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct inode *dir = dentry->d_parent->d_inode;
- struct autofs_sb_info *sbi = autofs4_sbi(dir->i_sb);
- int oz_mode = autofs4_oz_mode(sbi);
+ struct inode *dir;
+ struct autofs_sb_info *sbi;
+ int oz_mode;
int flags = nd ? nd->flags : 0;
int status = 1;
+ if (flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ dir = dentry->d_parent->d_inode;
+ sbi = autofs4_sbi(dir->i_sb);
+ oz_mode = autofs4_oz_mode(sbi);
+
/* Pending dentry */
spin_lock(&sbi->fs_lock);
if (autofs4_ispending(dentry)) {
*/
static int ceph_d_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct inode *dir = dentry->d_parent->d_inode;
+ struct inode *dir;
+
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ dir = dentry->d_parent->d_inode;
dout("d_revalidate %p '%.*s' inode %p offset %lld\n", dentry,
dentry->d_name.len, dentry->d_name.name, dentry->d_inode,
static int
cifs_d_revalidate(struct dentry *direntry, struct nameidata *nd)
{
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
if (direntry->d_inode) {
if (cifs_revalidate_dentry(direntry))
return 0;
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/spinlock.h>
+#include <linux/namei.h>
#include <asm/uaccess.h>
/* called when a cache lookup succeeds */
static int coda_dentry_revalidate(struct dentry *de, struct nameidata *nd)
{
- struct inode *inode = de->d_inode;
+ struct inode *inode;
struct coda_inode_info *cii;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ inode = de->d_inode;
if (!inode || coda_isroot(inode))
goto out;
if (is_bad_inode(inode))
*/
static int ecryptfs_d_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry);
- struct vfsmount *lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry);
+ struct dentry *lower_dentry;
+ struct vfsmount *lower_mnt;
struct dentry *dentry_save;
struct vfsmount *vfsmount_save;
int rc = 1;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ lower_dentry = ecryptfs_dentry_to_lower(dentry);
+ lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry);
if (!lower_dentry->d_op || !lower_dentry->d_op->d_revalidate)
goto out;
dentry_save = nd->path.dentry;
static int vfat_revalidate(struct dentry *dentry, struct nameidata *nd)
{
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
/* This is not negative dentry. Always valid. */
if (dentry->d_inode)
return 1;
static int vfat_revalidate_ci(struct dentry *dentry, struct nameidata *nd)
{
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
/*
* This is not negative dentry. Always valid.
*
*/
static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd)
{
- struct inode *inode = entry->d_inode;
+ struct inode *inode;
+
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+ inode = entry->d_inode;
if (inode && is_bad_inode(inode))
return 0;
else if (fuse_dentry_time(entry) < get_jiffies_64()) {
#include <linux/completion.h>
#include <linux/buffer_head.h>
#include <linux/gfs2_ondisk.h>
+#include <linux/namei.h>
#include <linux/crc32.h>
#include "gfs2.h"
static int gfs2_drevalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct dentry *parent = dget_parent(dentry);
- struct gfs2_sbd *sdp = GFS2_SB(parent->d_inode);
- struct gfs2_inode *dip = GFS2_I(parent->d_inode);
- struct inode *inode = dentry->d_inode;
+ struct dentry *parent;
+ struct gfs2_sbd *sdp;
+ struct gfs2_inode *dip;
+ struct inode *inode;
struct gfs2_holder d_gh;
struct gfs2_inode *ip = NULL;
int error;
int had_lock = 0;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ parent = dget_parent(dentry);
+ sdp = GFS2_SB(parent->d_inode);
+ dip = GFS2_I(parent->d_inode);
+ inode = dentry->d_inode;
+
if (inode) {
if (is_bad_inode(inode))
goto invalid;
* This file contains the code to do various system dependent things.
*/
+#include <linux/namei.h>
#include "hfs_fs.h"
/* dentry case-handling: just lowercase everything */
static int hfs_revalidate_dentry(struct dentry *dentry, struct nameidata *nd)
{
- struct inode *inode = dentry->d_inode;
+ struct inode *inode;
int diff;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ inode = dentry->d_inode;
if(!inode)
return 1;
static int jfs_ci_revalidate(struct dentry *dentry, struct nameidata *nd)
{
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
/*
* This is not negative dentry. Always valid.
*
fput(nd->intent.open.file);
}
+static int d_revalidate(struct dentry *dentry, struct nameidata *nd)
+{
+ int status;
+
+ status = dentry->d_op->d_revalidate(dentry, nd);
+ if (status == -ECHILD) {
+ if (nameidata_dentry_drop_rcu(nd, dentry))
+ return status;
+ status = dentry->d_op->d_revalidate(dentry, nd);
+ }
+
+ return status;
+}
+
static inline struct dentry *
do_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- int status = dentry->d_op->d_revalidate(dentry, nd);
+ int status;
+
+ status = d_revalidate(dentry, nd);
if (unlikely(status <= 0)) {
/*
* The dentry failed validation.
* the dentry otherwise d_revalidate is asking us
* to return a fail status.
*/
- if (!status) {
+ if (status < 0) {
+ /* If we're in rcu-walk, we don't have a ref */
+ if (!(nd->flags & LOOKUP_RCU))
+ dput(dentry);
+ dentry = ERR_PTR(status);
+
+ } else {
+ /* Don't d_invalidate in rcu-walk mode */
+ if (nameidata_dentry_drop_rcu_maybe(nd, dentry))
+ return ERR_PTR(-ECHILD);
if (!d_invalidate(dentry)) {
dput(dentry);
dentry = NULL;
}
- } else {
- dput(dentry);
- dentry = ERR_PTR(status);
}
}
return dentry;
if (!need_reval_dot(dentry))
return 0;
- status = dentry->d_op->d_revalidate(dentry, nd);
+ status = d_revalidate(dentry, nd);
if (status > 0)
return 0;
return -ECHILD;
nd->seq = seq;
- if (dentry->d_flags & DCACHE_OP_REVALIDATE) {
- /* We commonly drop rcu-walk here */
- if (nameidata_dentry_drop_rcu(nd, dentry))
- return -ECHILD;
+ if (dentry->d_flags & DCACHE_OP_REVALIDATE)
goto need_revalidate;
- }
path->mnt = mnt;
path->dentry = dentry;
__follow_mount_rcu(nd, path, inode);
* We may need to check the cached dentry for staleness.
*/
if (need_reval_dot(nd->path.dentry)) {
- if (nameidata_drop_rcu_maybe(nd))
- return -ECHILD;
- err = -ESTALE;
/* Note: we do not d_invalidate() */
- if (!nd->path.dentry->d_op->d_revalidate(
- nd->path.dentry, nd))
+ err = d_revalidate(nd->path.dentry, nd);
+ if (!err)
+ err = -ESTALE;
+ if (err < 0)
break;
}
return_base:
dir = nd->path.dentry;
case LAST_DOT:
if (need_reval_dot(dir)) {
- if (!dir->d_op->d_revalidate(dir, nd)) {
+ error = d_revalidate(nd->path.dentry, nd);
+ if (!error)
error = -ESTALE;
+ if (error < 0)
goto exit;
- }
}
/* fallthrough */
case LAST_ROOT:
#include <linux/kernel.h>
#include <linux/vmalloc.h>
#include <linux/mm.h>
+#include <linux/namei.h>
#include <asm/uaccess.h>
#include <asm/byteorder.h>
int res, val = 0, len;
__u8 __name[NCP_MAXPATHLEN + 1];
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
parent = dget_parent(dentry);
dir = parent->d_inode;
#include <linux/vfs.h>
#include <linux/mount.h>
#include <linux/seq_file.h>
+#include <linux/namei.h>
#include <linux/ncp_fs.h>
* component of the path.
* We check for this using LOOKUP_CONTINUE and LOOKUP_PARENT.
*/
-static inline unsigned int nfs_lookup_check_intent(struct nameidata *nd, unsigned int mask)
+static inline unsigned int nfs_lookup_check_intent(struct nameidata *nd,
+ unsigned int mask)
{
if (nd->flags & (LOOKUP_CONTINUE|LOOKUP_PARENT))
return 0;
* If the parent directory is seen to have changed, we throw out the
* cached dentry and do a new lookup.
*/
-static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd)
+static int nfs_lookup_revalidate(struct dentry *dentry, struct nameidata *nd)
{
struct inode *dir;
struct inode *inode;
struct nfs_fattr *fattr = NULL;
int error;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
parent = dget_parent(dentry);
dir = parent->d_inode;
nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE);
static int ocfs2_dentry_revalidate(struct dentry *dentry,
struct nameidata *nd)
{
- struct inode *inode = dentry->d_inode;
+ struct inode *inode;
int ret = 0; /* if all else fails, just return false */
- struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb);
+ struct ocfs2_super *osb;
+
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ inode = dentry->d_inode;
+ osb = OCFS2_SB(dentry->d_sb);
mlog_entry("(0x%p, '%.*s')\n", dentry,
dentry->d_name.len, dentry->d_name.name);
*/
static int pid_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct inode *inode = dentry->d_inode;
- struct task_struct *task = get_proc_task(inode);
+ struct inode *inode;
+ struct task_struct *task;
const struct cred *cred;
+ if (nd && nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ inode = dentry->d_inode;
+ task = get_proc_task(inode);
+
if (task) {
if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) ||
task_dumpable(task)) {
static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct inode *inode = dentry->d_inode;
- struct task_struct *task = get_proc_task(inode);
- int fd = proc_fd(inode);
+ struct inode *inode;
+ struct task_struct *task;
+ int fd;
struct files_struct *files;
const struct cred *cred;
+ if (nd && nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ inode = dentry->d_inode;
+ task = get_proc_task(inode);
+ fd = proc_fd(inode);
+
if (task) {
files = get_files_struct(task);
if (files) {
*/
static int proc_base_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct inode *inode = dentry->d_inode;
- struct task_struct *task = get_proc_task(inode);
+ struct inode *inode;
+ struct task_struct *task;
+
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ inode = dentry->d_inode;
+ task = get_proc_task(inode);
if (task) {
put_task_struct(task);
return 1;
#include <linux/sysctl.h>
#include <linux/proc_fs.h>
#include <linux/security.h>
+#include <linux/namei.h>
#include "internal.h"
static const struct dentry_operations proc_sys_dentry_operations;
static int proc_sys_revalidate(struct dentry *dentry, struct nameidata *nd)
{
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
return !PROC_I(dentry->d_inode)->sysctl->unregistering;
}
static int xattr_hide_revalidate(struct dentry *dentry, struct nameidata *nd)
{
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
return -EPERM;
}
static int sysfs_dentry_revalidate(struct dentry *dentry, struct nameidata *nd)
{
- struct sysfs_dirent *sd = dentry->d_fsdata;
+ struct sysfs_dirent *sd;
int is_dir;
+ if (nd->flags & LOOKUP_RCU)
+ return -ECHILD;
+
+ sd = dentry->d_fsdata;
mutex_lock(&sysfs_mutex);
/* The sysfs dirent has been deleted */
#define DCACHE_OP_REVALIDATE 0x4000
#define DCACHE_OP_DELETE 0x8000
-
extern spinlock_t dcache_inode_lock;
extern seqlock_t rename_lock;
git.openpandora.org / pandora-kernel.git / commitdiff